CN101324939A - System and method for forecasting new business market based on data development - Google Patents
System and method for forecasting new business market based on data development Download PDFInfo
- Publication number
- CN101324939A CN101324939A CNA2008101350435A CN200810135043A CN101324939A CN 101324939 A CN101324939 A CN 101324939A CN A2008101350435 A CNA2008101350435 A CN A2008101350435A CN 200810135043 A CN200810135043 A CN 200810135043A CN 101324939 A CN101324939 A CN 101324939A
- Authority
- CN
- China
- Prior art keywords
- data
- business
- new
- prediction
- empirical data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a system for predicting the new business market based on data mining and the method thereof. The method is used for predicting the number of clients and profit of a new product or a new business through the existing products, business and the related client resource data. The method comprises the following steps: (1) determining industry latitude according to the characteristics of different industries; (2) collecting data of the original business according to the required latitude data; (3) calculating and predicting according to the collected data to obtain the result of analysis and prediction; and (4) adding empirical data adjustment parameters and rectifying empirical data through the parameters in the process of calculation and analysis, thereby improving the precision of prediction. The system comprises four parts of an industry latitude determining module through parameters, a data collecting module, a prediction module and an empirical data adjustment parameter module. The method can be used for predicting the new business in various industries, so that enterprises data can support the new business decisions.
Description
Technical field:
The present invention relates to data mining field and computer aided decision making field, also can be described as the application of data mining in the computer aided decision making field.Be a kind ofly to predict new product or the professional client's quantity and the method for income specifically by existing product, business and relevant customer resources data.
Background technology:
Current era is the epoch of infotech, most of enterprise has finished the informationization of operation control, particularly in more enterprises of application message technology such as bank, insurance, telecommunications, the client resources management system that oneself has all been arranged, but the data of collecting often can not effectively be analyzed, and can not be used for the market outlook prediction of new product.Utilize data mining like this, new product in the design or new business are carried out market forecast demand has just been arranged.
Aspect market forecast, generally use following two kinds of theories at present:
Time series forecasting, in market forecast, often run into a series of economic target values that change according to the time, sales volume, the consumer in (season) take in over the years per year, purchasing power increases statistical value etc. as enterprise's product, and these one group of data of in chronological sequence lining up are called time series.Carry out forecast method according to time series and be called time series forecasting.
Regression Forecast, in economic projection, people forecasting object (economic target) as dependent variable, those and the closely-related influence factor of forecasting object as independent variable.According to history of the two and present statistical data, set up regression model, through being used for prediction after the statistical test.
These two kinds of methods all concentrate on the main trend of analyzing market market forecast, the general lasting macro-indicators that exists of analysis that is fit to.Prediction to new product new business is felt inadequate.
The present invention is applicable to different industries, carries out market forecast according to the similarity of new business and old business, provided computing formula, and it is less influenced by the operational staff.Especially, this method is adjusted parameters of formula by the historical statistics data, can better adapt to the characteristics of industry-by-industry.
Summary of the invention:
In order to find a kind of new business forecast method that can use at industry-by-industry, overcome the obstacle of existing technologies, make the data of enterprise really produce supporting role to the new business decision-making, the present invention begins one's study and finishes.
System is divided into that parameter is determined industry dimension, data aggregation, prediction, rule of thumb data are adjusted four modules of parameter.
Parameter is determined the industry dimension: by definition and similar dimension to similar dimension industrial characteristic is determined in the influence of new old service similarity coefficient.For example: can use expense type, consumption amount, preferential ratio etc. as dimension in telecommunications industry.
Data aggregation: carry out the collection of original business datum according to required dimension data, and collect original business (or product) client amount and business income.
Prediction: the core of implementation method model, data that responsible basis is collected and empirical data, parameter are calculated and are estimated client's quantity, income.This part is used following formula:
λ=∑(x1×y1×z1,...,xn×yn×zn)×t+λs
λ representative prediction client amount; X is old service client amount; Y is the likelihood (the different industries different business has different likelihood algorithms) of the new old service of calculating according to dimension; Z is price influence coefficient, z=old product price/new product price; T is that market forecast trend coefficient is not (when multiply by t, be to calculate gained according to old service, partly select the user of old service to select new business, multiply by t and considered market forecast trend, in concrete industry, can obtain different values according to the macroeconomy data, when increasing fast such as the industry whole market, this number is the constant greater than 1); λ s is the empirical data modified value.
ε=λ×(∑(m1×x1×y1×z1,…,mn×xn×yn×zn)/(∑(x1×y1×z1,…,xn×yn×zn)×p+εs)
ε is the prediction income; λ is the prediction client amount in the following formula; M is every client's average yield of old service; X, y, z are the same; P is the coefficient of ratio between new business price and old service are on average fixed a price, the mean value of p=new product price/old product price; ε s is the empirical data modified value of income.
Empirical data is adjusted parameter: by historical tentative calculation and real data, predict the outcome and actual generation result's statistical study, obtain above-mentioned two empirical data modified values.
The present invention is a point of penetration with the similarity of new old service, meet the market discipline that new business is released, and considered the influence of price factor when new business is released, can accurately predict the market reaction of new business, and can improve accuracy of predicting by the empirical data correction, for maximum support has been carried out in the release decision-making of new business (or product).
Description of drawings:
Fig. 1: system module process flow diagram
Embodiment:
The system module process flow diagram as shown in Figure 1, and is existing with the example that is embodied as in telecommunications industry, and the present invention is further elaborated.
New business---broadband (4M) timing is 10 yuan customer volume and income per hour in prediction.
Step 1: dimension is defined as fee item (referring to local telephone network, long-distance etc. classification), type of service (fixed line, Personal Handyphone System, mobile phone, broadband, WAP etc.), charge type, price, average cost.The ratio that each dimension influences similarity coefficient is defined as 0.3,0.2,0.1,0.2,0.2 successively.
Step 2: obtain each dimension data of old service and client's number, income by telecommunication charging system and client resources management system.There are three old service data related in the database with new business.As shown in the table:
Professional name | Fee item | Type of service | Charge type | Price | Average cost | Client's number | Income (unit) |
The 512K timing | ADSL takes | ADSL512K | Timing | 1 yuan/hour | 50 | 613414 | 30670700 |
The 1M timing | ADSL takes | ASDL1M | Timing | 2.5 unit/hour | 120 | 34165 | 4099800 |
The 2M timing | ADSL takes | ADSL2M | Timing | 5 yuan/hour | 150 | 53002 | 7950300 |
Step 3: calculate the similarity of each old service and new business according to each dimension data of new business, be followed successively by 0.31,0.52,0.78; The price influence coefficient is 0.10,0.25,0.31; Obtaining the market trend coefficient in the telecommunications industry is 1.13; Two empirical data modified values are 0.Bring core formula prediction client's amount and income then into.
λ=∑(x1×y1×z1,...,xn×yn×zn)×t+λs
=(613414×0.31×0.10+34165×0.52×0.25+53002×0.78×0.31)×1.13+0
≈40988
ε=λ×(∑(m1×x1×y1×z1,…,mn×xn×yn×zn)/(∑(x1×y1×z1,…,xn×yn×zn)×p+εs)
=40988×((50×613414×0.31×0.10+120×34165×0.52×0.25+150×53002×0.78×0.31)/(613414×0.31×0.10+34165×0.52×0.25+53002×0.78×0.31)×10/(1+2.5+5)+0)
≈13584295
For example: obtaining client's amount is 40988, and income is 13584295.
Step 4: after the prediction, continue to collect new business and release the back real achievement, adjust empirical data by the predicted value and the actual value that have obtained.When predicting again, new empirical data will be used.
The above; it only is an example of embodiment of the present invention; can not therefore limit to interest field of the present invention; concerning those of ordinary skill in the art; all utilizations technical scheme of the present invention and technical conceive are made other change and distortion, all should belong within the protection domain of claim of the present invention.
Claims (3)
1, a kind of new business market forecast system based on data mining is characterized in that system is divided into following four modules:
Parameter is determined industry dimension module: by definition and similar dimension to similar dimension industrial characteristic is determined in the influence of new old service similarity coefficient; For example: can use expense type, consumption amount, preferential ratio etc. as dimension in telecommunications industry;
Data collection module: carry out the collection of original business datum according to required dimension data, and collect original business (or product) client amount and business income;
Prediction module: the core of implementation method model, data that responsible basis is collected and empirical data, parameter are calculated and are estimated client's quantity, income;
Empirical data is adjusted parameter module: by historical tentative calculation and real data, predict the outcome and actual generation result's statistical study, obtain the empirical data modified value.
2, a kind of new business market forecast method based on data mining is characterized in that specifically may further comprise the steps:
A, determine industry dimension parameter,, define the similar latitude of the sector, and determine of the influence of similar latitude for the similarity coefficient of new old service according to the industrial characteristic of different industries;
B, data aggregation according to required latitude data, are collected original business datum, such as: charge type, service order amount, business income etc.
C, calculating, prediction.According to the data of collecting,, calculate the expectation client quantity and the estimated revenue of new business by the computing formula of core;
D, empirical data is set adjusts parameter, by this parameter is set and revises empirical data, thereby improve accuracy of predicting.
3,, it is characterized in that the core calculations formula of C, D step and empirical data adjustment parameter are provided with as follows as claim 2 kind of described a kind of new business Forecasting Methodology based on data mining:
λ=∑(x1×y1×z1,...,xn×yn×zn)×t+λs
λ representative prediction client amount; X is old service client amount; Y is the likelihood (the different industries different business has different likelihood algorithms) of the new old service of calculating according to dimension; Z is price influence coefficient, z=old product price/new product price; T is that market forecast trend coefficient is not (when multiply by t, be to calculate gained according to old service, partly select the user of old service to select new business, multiply by t and considered market forecast trend, in concrete industry, can obtain different values according to the macroeconomy data, when increasing fast such as the industry whole market, this number is the constant greater than 1); λ s is the empirical data modified value;
ε=λ×(∑(m1×x1×y1×z1,…,mn×xn×yn×zn)/(∑(x1×y1×z1,…,xn×yn×zn)×p+εs)
ε is the prediction income; λ is the prediction client amount in the following formula; M is every client's average yield of old service; X, y, z are the same; P is the coefficient of ratio between new business price and old service are on average fixed a price, the mean value of p=new product price/old product price; ε s is the empirical data modified value of income;
Empirical data is adjusted parameter: by historical tentative calculation and real data, predict the outcome and actual generation result's statistical study, obtain above-mentioned two empirical data modified values.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2008101350435A CN101324939A (en) | 2008-07-29 | 2008-07-29 | System and method for forecasting new business market based on data development |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2008101350435A CN101324939A (en) | 2008-07-29 | 2008-07-29 | System and method for forecasting new business market based on data development |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101324939A true CN101324939A (en) | 2008-12-17 |
Family
ID=40188470
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA2008101350435A Pending CN101324939A (en) | 2008-07-29 | 2008-07-29 | System and method for forecasting new business market based on data development |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101324939A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102622698A (en) * | 2012-02-17 | 2012-08-01 | 内蒙古东部电力有限公司 | Electricity market analyzing and predicting system and analyzing and predicting method thereof |
CN104573108A (en) * | 2015-01-30 | 2015-04-29 | 联想(北京)有限公司 | Information processing method and information processing unit |
CN105303260A (en) * | 2015-11-06 | 2016-02-03 | 华中科技大学 | Impact prediction method and system of new business of mobile communication |
CN105493126A (en) * | 2014-01-14 | 2016-04-13 | 秀投徐富粦株式会社 | Card benefit providing system and method for automatically adjusting card benefits for each customer by applying card usage information |
CN106204098A (en) * | 2016-06-28 | 2016-12-07 | 郑州师范学院 | A kind of whole world underwear industry data is collected and analysis platform and the method for analysis thereof |
CN108764553A (en) * | 2018-05-21 | 2018-11-06 | 世纪龙信息网络有限责任公司 | Userbase prediction technique, device and computer equipment |
-
2008
- 2008-07-29 CN CNA2008101350435A patent/CN101324939A/en active Pending
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102622698A (en) * | 2012-02-17 | 2012-08-01 | 内蒙古东部电力有限公司 | Electricity market analyzing and predicting system and analyzing and predicting method thereof |
CN105493126A (en) * | 2014-01-14 | 2016-04-13 | 秀投徐富粦株式会社 | Card benefit providing system and method for automatically adjusting card benefits for each customer by applying card usage information |
CN104573108A (en) * | 2015-01-30 | 2015-04-29 | 联想(北京)有限公司 | Information processing method and information processing unit |
CN105303260A (en) * | 2015-11-06 | 2016-02-03 | 华中科技大学 | Impact prediction method and system of new business of mobile communication |
CN106204098A (en) * | 2016-06-28 | 2016-12-07 | 郑州师范学院 | A kind of whole world underwear industry data is collected and analysis platform and the method for analysis thereof |
CN108764553A (en) * | 2018-05-21 | 2018-11-06 | 世纪龙信息网络有限责任公司 | Userbase prediction technique, device and computer equipment |
CN108764553B (en) * | 2018-05-21 | 2020-12-15 | 世纪龙信息网络有限责任公司 | User scale prediction method and device and computer equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11823225B2 (en) | Methods and apparatus to incorporate saturation effects into marketing mix models | |
CN109428344B (en) | Multi-power-supply investment planning method and device comprising wind power plant | |
US20040215529A1 (en) | System and method for energy price forecasting automation | |
CN111080101A (en) | AHP-based multi-dimensional evaluation method for service efficiency of power supply channel | |
CN101324939A (en) | System and method for forecasting new business market based on data development | |
US20120203596A1 (en) | Demand side management portfolio manager system | |
CN110991875A (en) | Platform user quality evaluation system | |
Peña-Guzmán et al. | Forecasting Water Demand in Residential, Commercial, and Industrial Zones in Bogotá, Colombia, Using Least‐Squares Support Vector Machines | |
CN107818386A (en) | Power grid enterprises' Operating profit Forecasting Methodology | |
CN106296247A (en) | The online sort method of network information resource and device | |
Tanaka et al. | Spatial spillovers from foreign direct investment: Evidence from the Yangtze River Delta in China | |
CN113256324A (en) | Data asset pricing method, device, computer equipment and storage medium | |
Fu et al. | Bike-sharing inventory management for market expansion | |
CN110717666A (en) | Market competitiveness evaluation method for enterprise products | |
JPH05314094A (en) | Sales prediction supporting device | |
Franta et al. | Forecasting Czech GDP using mixed-frequency data models | |
CN114254916A (en) | Multi-dimensional supplier risk real-time early warning method with power grid service characteristics | |
JPWO2021260678A5 (en) | ||
Thuring | A credibility method for profitable cross-selling of insurance products | |
CN106803194A (en) | Online competitive price probabilistic model generation method | |
CN104778636B (en) | Method is recommended in the information service of apparel maintenance support | |
CN104978604B (en) | A kind of analog simulation method and device based on professional ability model | |
CN113554468A (en) | Market saturation-based benediction analysis method | |
Xing et al. | The evaluation of customer potential value based on prediction and cluster analysis | |
Azoev et al. | Benchmarking of market capacity assessment methods and their application to digital products |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20081217 |